IVP COURSE PROJECT

“COVID-19 detection in Chest X-ray

images using deep learning techniques”
C-3
Group NO. - 10

Presented By -
NIKHIL DUBEY (IIT2020016)
Raunak Krishan Jaiswal (IIT2020031)
Aditya Biswakarma (IIT2020033)
Anupam raj (IIT2020034)
Katam Bala Prasanna Babu (IIT2020040)
Jay Suthar (IIT2020087)
Rohan Tirkey (IIT2020088)

Instructor - Prof. Anupam Agarwal

 Index
1. Problem Statement
2. Introduction
3. Literature review
4. Methodology & flowchart
5. Experimental data descriptions
6. Tools, Language & libraries used for
implementation
7. Activity chart
8. Intermediate result snapshots with their description
9. Work remaining
10.References
2
 Problem Statement
The objective is to create a model which
detects the
COVID-19 Chest X-ray images using deep learning techniques.

3
 Introduction
The year 2020 was marked by a pandemic disease caused by a type of the coronavirus (CoV)
family, called COVID-19 or SARS-CoV-2, which led to more than four million infections and
more than 355,000 deaths worldwide.

This research paper has created an accurate classification method to perform reliable detection
of the viral COVID-19 patterns. we have proposed a Deep Convolutional Neural Network-
based solution which can detect the COVID-19 +ve patients using chest X-Ray images.

We have proposed a solution based on a deep convolutional neural network that can detect
patients with COVID-19+ using chest X-rays. Several state-of-the-art CNN models – VGG16,
VGG19, and ResNet50 – were adopted in the proposed work. We will also implement a custom
CNN architecture model. These models will be individually trained to make independent
predictions. And at the end, we compare the results for each approach for accuracy.

The widespread use of CNNs for image classification tasks is due to the fact that they have
demonstrated high accuracy in the areas of image recognition and object detection.
4
 Literature review
We have used 8 research papers for our reference -

The Complete LR can be accessed with the following link -

https://docs.google.com/spreadsheets/d/1LQwwrCzTIFYpspgto2lBXqFJiUETaC6q/edit?usp=sharing&
ouid=102962082556326646196&rtpof=true&sd=true

5
 Methodology & flowchart
Convolutional Neural Network (CNN)- A Convolutional Neural Network (ConvNet/CNN) is
a Deep Learning algorithm that can take in an input image, assign importance (learnable weights and
biases) to various aspects/objects in the image, and be able to differentiate one from the other. The pre-
processing required in a ConvNet is much lower as compared to other classification algorithms. While
in primitive methods filters are hand-engineered, with enough training, ConvNets have the ability to
learn these filters/characteristics.

Like other neural networks, a CNN is composed of an input layer, an output layer, and many hidden
layers in between.

6
 The proposed CNN is based on pre-trained transfer models such as VGG16, VGG19, and ResNet50
as well as our own architectural model in order to obtain high prediction accuracy from a small
sample of X-ray images. The images are classified into two classes, No_findings/normal and Covid-
19.

To overcome the insufficient data and training time, a transfer learning technique is applied by
employing the ImageNet dataset. The implementation will use Keras with a TensorFlow backend,
which will be performed, then adapted to our dataset full of X-ray images in Covid-19 and
No_findings folders.
7
 VGG16 -

VGG19 -

8
 ResNet50

Custom Architecture(Own Architecture)

9
 Experimental data descriptions
We have taken the dataset for this model from
https://github.com/ieee8023/covid-chestxray-dataset .

This Data is collected from public sources as well as through indirect

collection from hospitals and physicians.

Data impact: Image data linked with clinically relevant attributes in a public
dataset that is designed for ML will enable parallel development of these
tools and rapid local validation of models.

10
 Tools, Language & libraries used for
implementation
Software Technologies Required -
● Python (latest version)
● Or Google Colab Or Jupyter Notebook for running ipynb file
● Other important libraries required will be -
○ TensorFlow
○ Keras
○ Scikit-Learn (sklearn)
○ Pandas
○ Numpy
○ Matplotlib (pyplot)
○ seaborn

Hardware Requirements
● Graphics Card / GPU (if possible) (Google Colab will perform better)

11
 Activity Time Chart

12
 Intermediate result snapshots with
their description
ResNet50

train/validation Accuracy and Loss graphs - Confusion Matrix

— Please note that 0 and 1 values represent
13 Covid-19 and No_findigs respectively.
 test of our model against more random
samples from the test data -

14
 VGG16

train/validation Accuracy and Loss graphs -

Confusion Matrix
— Please note that 0 and 1 values represent
Covid-19 and No_findigs respectively.
15
 test of our model against more random
samples from the test data -

16
 VGG19

train/validation Accuracy and Loss graphs -

Confusion Matrix
— Please note that 0 and 1 values represent
Covid-19 and No_findigs respectively.
17
 test of our model against more random
samples from the test data -

18
 Custom(Own Model)

train/validation Accuracy and Loss graphs -

Confusion Matrix
— Please note that 0 and 1 values represent
Covid-19 and No_findigs respectively.
19
 test of our model against more random
samples from the test data -

20
 Work Remaining
We need to do more testing of this model for evaluating the performance of a fully
trained model on a testing set.

We need to use more dataset for hyperparameter tuning for maximizing the model's
performance, minimizing a predefined loss function to produce better results with
fewer errors.

21
 References
Ahsan, Mominul, Md Abdul Based, Julfikar Haider, and Marcin Kowalski. "COVID-19 detection from chest X-
ray images using feature fusion and deep learning." Sensors 21, no. 4 (2021): 1480.

Das, Amit Kumar, Sayantani Ghosh, Samiruddin Thunder, Rohit Dutta, Sachin Agarwal, and Amlan Chakrabarti.
"Automatic COVID-19 detection from X-ray images using ensemble learning with convolutional neural network."
Pattern Analysis and Applications 24, no. 3 (2021): 1111-1124.

Chen, Joy Iong-Zong. "Design of accurate classification of COVID-19 disease in X-ray images using deep
learning approach." Journal of ISMAC 3, no. 02(2021): 132-148.

Makris, Antonios, Ioannis Kontopoulos, and Konstantinos Tserpes. "COVID-19 detection from chest X-Ray
images using Deep Learning and Convolutional Neural Networks." In 11th hellenic conference on artificial
intelligence, pp. 60-66. 2020.

22
 References(Contd.)
Hussain, Emtiaz, Mahmudul Hasan, Md Anisur Rahman, Ickjai Lee, Tasmi Tamanna, and Mohammad Zavid
Parvez. "CoroDet: A deeplearning based classification for COVID-19 detection using chest X-ray images." Chaos,
Solitons & Fractals 142 (2021): 110495.

Akter, Shamima, FM Javed Mehedi Shamrat, Sovon Chakraborty, Asif Karim, and Sami Azam. "COVID-19
detection using deep learning algorithm on chest X-ray images." Biology 0, no. 11 (2021): 1174.

Yang, Dandi, Cristhian Martinez, Lara Visuña, Hardev Khandhar, Chintan Bhatt, and Jesus Carretero.
"Detection and analysis of COVID-19 in medical images using deep learning techniques." Scientific Reports 11, no.
1 (2021): 1-13.

Reshi, Aijaz Ahmad, Furqan Rustam, Arif Mehmood, Abdulaziz Alhossan, Ziyad Alrabiah, Ajaz Ahmad, Hessa
Alsuwailem, and Gyu Sang Choi. "An efficient CNN model for COVID-19 disease detection based on X-ray image
23 classification." Complexity 2021 (2021).
 THANKS!

24